【摘要】 针对板带材轧制是一个复杂的非线性过程 ,板形控制 (AFC)和板厚控制 (AGC)又是相互耦合的一个综合系统等特点 ,该文首先采用神经网络分散解耦方法 ,对此板形板厚多变量耦合系统进行解耦 ,而后再应用基于免疫遗传算法的PID控制对解耦后的已近似成为两个独立的SISO系统的广义对象进行控制 ,从而建立了基于免疫遗传算法的板形板厚神经网络分散解耦PID控制系统。仿真结果证明了此AFC -AGC控制系统具有良好的自适应跟随和抗扰性能 ,其控制效果优于传统的解耦PID控制。更多还原

【Abstract】 In the paper, because strip rolling is a very complicated nonlinear process, and characterized by couple between automatic flatness control (AFC) and automatic gauge control(AGC), the authors firstly make this coupling multivariable system decouple using neural network decentralized-decoupling method. Secondly, authors adopt PID controller based on immune genetic algorithm to control the decoupled generalized object which can be approximately equal to two SISO systems. Thus, construct strip flatness and gauge neural network decentralized-decoupling PID control system based on immune genetic algorithm and simulate on it. The simulation results show that this kind of new controller has good performances of adaptively tracking target and resisting disturbances and is superior to the conventional decoupled PID control in terms of improving the strip flatness and gauge accuracy.更多还原